An Improved hybrid Salp Swarm Optimization and African Vulture Optimization Algorithm for Global Optimization Problems and Its Applications in Stock Market Prediction

Abstract

Optimization is necessary for solving and improving the solution of various complex problems. Every meta-heuristic algorithm can have a weak point, and multiple mechanisms and methods can be used to overcome these weaknesses. We can use hybrid meta-heuristic algorithms to arrive at an efficient algorithm. This paper presents a new and intelligent approach by hybridizing meta-heuristic algorithms and using different mechanisms simultaneously without significantly increasing the time complexity. For this purpose, two algorithms, Salp Swarm Optimization(SSO) and the African Vulture Optimization Algorithm (AVOA) have been hybridized. And to improve the optimization process of the Modified Choice Function and Learning Automata mechanisms. In addition, two other improving mechanisms, named Opposition-Based Learning (OBL) and β-hill climbing (BHC) technique, have been presented and integrated with the AVOA-SSA algorithm. Fifty-two standard benchmarks were used to test and evaluate the AVOA-SSA algorithm. Finally, an improved version of the Extreme Learning Machine(ELM) classifier has been used with real stock market data for stock market prediction. The obtained results indicate the excellent and acceptable performance of the AVOA-SSA algorithm in `solving optimization problems and has been able to achieve high-quality solutions.